EVALUATION OF BIOMARKERS IN CHRONIC INFLAMMATORY DISEASES

From the Institute of Environmental Medicine Karolinska Institutet, Stockholm, Sweden

EVALUATION OF BIOMARKERS IN CHRONIC INFLAMMATORY DISEASES

Zahra Golabkesh

Stockholm 2020

All previously published papers were reproduced with permission from the publisher.

Published by Karolinska Institutet.

Printed by Arkitektkopia AB, 2020

Evaluation of biomarkers in chronic inflammatory diseases

THESIS FOR DOCTORAL DEGREE (Ph.D.)

By

Zahra Golabkesh

Principal Supervisor:

Assistant Professor Boel Brynedal, PhD Karolinska Institutet

Institute of Environmental Medicine Co-supervisors:

Associate Professor Karin Leander, PhD Karolinska Institutet

Institute of Environmental Medicine Division of Cardiovascular and Nutrition Epidemiology

Professor Lars Alfredsson, PhD Karolinska Institutet

Institute of Environmental Medicine Division of Occupational Medicine Assistant Professor Paolo Frumento, PhD University of Pisa

Department of Political Sciences

Opponent:

Professor Majken Karoline Jensen, PhD University of Copenhagen

Division of Epidemiology Examination Board:

Professor Magnus Bäck, MD PhD Karolinska University Hospital Department of Medicine

Division of Cardiovascular Medicine Associate Professor Meliha Kapetanovic, MD PhD

Lund University

Department of Clinical Sciences Division of Rheumatology

Associate Professor Olof Gidlöf, PhD Lund University

Department of Clinical Sciences Division of Molecular Cardiology

To Behrouz and Nova &

To those who do not surrender in difficulties

"I don't have any particular recipe . . . Doing research is challenging as well as attractive. It is like being lost in a jungle and trying to use all the knowledge that you can gather to come up with some new tricks, and with some luck you might find a way out."

ABSTRACT

The overall aim of this thesis was to assess the role of biomarkers in chronic inflammatory disease. The goal of the first study was to investigate the associa- tion of sIL-6R and sgp130 levels and risk of first myocardial infarction (MI), and to explore the potential interaction between these two biomarkers in association with risk of MI.

In the second study, the aim was to identify genetic loci associated with the circu- lation of sgp130, known as an inhibitor of the IL-6 trans-signaling, and also to test the association between the single nucleotide polymorphisms (SNPs) associated with sgp130 and carotid intima-media thickness (c-IMT) as a surrogate marker of atherosclerosis.

The main objective of the third study was to investigate whether environmental risk factors of rheumatoid arthritis (RA) affect the occurrence of different reac- tivity patterns to citrullinated peptides and the presence of rheumatoid factor (RF) isotypes.

Study I was carried out in the Stockholm Heart Epidemiology Program (SHEEP), a case-control study designed to determine new risk factors. sIL-6R and sgp130 levels were measured in serum of 682 and 664 MI cases and 1103 and 1062 con- trols, respectively. Increased concentrations of sIL6R were associated with a high occurrence of MI while very high levels of sgp130 had an inverse association with incidence of MI. The obtained results indicated presence of an interaction between the low level of sgp130 and the high level of sIL6R that might have a synergistic effect on increased risk of MI. Our results highlighted the necessity of focusing on molecular pathways instead of only one biomarker when estimating the risk of CVDs.

Study II was performed in IMPROVE, a European multicenter prospective study recruited subjects with high risk profile for cardiovascular diseases. c-IMT was measured at baseline and after 30 months follow-up. Genomic DNA of 3703 par- ticipants was genotyped using the CardioMetaboChip and ImmunoChip. A linear regression model was used to assess the association of 360,842 SNPs and sgp130 levels. Model was adjusted for gender, age and population structure. rs10935473 and rs1929666, located at chromosome 3 and 10 respectively, had an association with sgp130 circulation. Besides, 24 SNPs showed the suggestive association with sgp130 levels. A positive association between rs17688225 and serum level of sgp130 was observed while this SNP showed a negative association with c-IMT.

Our results indicated that sgp130 levels is regulated by multiple genetic loci, which

In study III, analyses were based on data from two Swedish RA patients cohorts:

Epidemiological Investigation of RA (EIRA, n=2859), and early RA from Umeå (eRA-Umeå, n=1011). Our results corroborate the wide-spread co-occurrence of different RA-specific antibodies, which is likely mediated by epitope spreading or cross-reactivity. When analyzing all antibodies jointly, we found that smoking is mainly associated with the presence of anti-Cit-Fibα36-50, anti-CEP-1 as well as with IgA-RF. No conclusive associations were found between low alcohol consumption or high BMI with the presence of any specific autoantibody. Our study indicated smoking might play a part in pathogenesis of RA through specific anticitrulline immunity by increasing the exposure of these antigens in affected tissues like lungs.

SAMMANFATTNING PÅ SVENSKA

Det övergripande syftet med denna avhandling var att utvärdera biomarkörernas roll i kronisk inflammatorisk sjukdom. Målet med den första studien var att undersöka sambandet mellan sIL-6R- och sgp130-nivåerna och risken för en första episod av hjärtinfarkt, samt undersöka den potentiella interaktionen mellan dessa två biomarkörer i risk för hjärtinfarkt. I den andra studien syftade vi till att identifiera genetiska varianter kopplade till sgp130-serumnivåer, den naturliga antagonisten mot IL6-transignaleringen. Vi undersökte även sambandet mellan genotyperna förknippade med sgp130 och en markör för subklinisk åderförkalkning (carotis intima-media tjocklek, c-IMT). Målet med den tredje studien var att undersöka om miljöfaktorer för reumatoid artrit (RA) påverkar förekomsten av sjukdomsspeci- fika antikropparna mot citrullerade peptider och reumatoid faktor (RF) isotyper.

Studie I genomfördes i en stor befolkningsbaserad fallkontrollstudie, Stockholm Heart Epidemiology Program (SHEEP). Vi mätte nivåer av sIL6R och sgp130 i serumprover från 682 respektive 664 hjärtinfarktspatienter, och 1103 respektive 1062 kontroller. Förhöjda koncentrationer av sIL6R var förknippade med en ökad förekomst av hjärtinfarkt och mycket höga nivåer av sgp130 var associerade med en reducerad förekomst av hjärtinfarkt. Våra resultat indikerar även en interaktion mellan dessa två biomarkörer, som föreslår att låga sgp130-nivåer tillsammans med höga sIL6R-nivåer öka risken för hjärtinfarkt mer än förväntat.

Studie II utfördes i IMPROVE, en europeisk prospektiv multicenterstudie utformad för att undersöka sambandet mellan c-IMT och c-IMT-progression med risken för framtida kardiovaskulära sjukdomar. Genomiskt DNA genotypades med hjälp av CardioMetaboChip och ImmunoChip. rs10935473 (på kromosom 3) och rs1929666 (på kromosom 10) var signifikant förknippade med sgp130-nivåer och 24 genetiska varianter visade svagare association med sgp130-nivåer. En av de genetiska vari- anterna med svagare association, rs17688225 på kromosom 14, visade en negativ association med c-IMT. Våra resultat indikerar att sgp130-nivåerna regleras av flera genetiska varianter som till viss del överlappar dem som reglerar c-IMT.

Studie III använde data från två svenska kohorter av RA-patienter: Epidemiologisk undersökning av RA (EIRA, n = 2859) och tidig RA från Umeå (eRA-Umeå, n = 1011). Våra resultat bekräftar att patienter oftast utrycker flera olika antikroppar som binder flera olika auto-antigen samtidigt. Detta medieras troligen av epitop- spridning eller korsreaktivitet. När vi analyserade alla antikroppar gemensamt fann vi att rökning huvudsakligen är förknippat med närvaron av anti-Cit-Fibα36-50, anti-CEP-1 såväl som med IgA-RF. Vi hittade inga tydliga samband mellan låg alkoholkonsumtion eller hög BMI med närvaron av någon specifik autoantikropp.

یرتکد عطقم همان ناياپ هصلاخ

فادھا یتسيز یاھ صخاش شقن یسررب همان ناياپ نيا یلصا فدھ (Biomarkers)

تسا یباھتلا نمزم یاھ یراميب رد

.

هعلاطم فادھا لوا

تسا رارق نيا زا همان ناياپ نيا

: . ١

هرامش نيکولرتنيا هدنريگ لولحم لکش نازيم ی هطبار یسررب ۶

(SIL6R) ی هرامش نيئتورپ وکيلگ لولحم لکش و ١٣٠

(SGP130)

اب .یبلق هتکس نيلوا عوقو رطخ

٢ . یبلق هتکس رطخ اب هطبار رد یتسيز صخاش ود نيا نيب یلامتحا شنکمھرب ناکما یسررب .

ام همان ناياپ نيا مود هعلاطم رد لگ حطس اب طبترم یکيتنژ یاھ هيور عاونا ات مينآرب

ي هرامش نيئتورپ وک ١٣٠

ميبايب ار ناسنا نوخرد

. ئتورپ وکيلگ نژ یديتوئلکون کت یتخير دنچ عون زا یناسنا تيعمج یکيتنژ یاھ ینوگانوگ نيب طابترا ات ميدرک ششوک نينچمھ ی هرامش ني

١٣٠ (SGP130) ماميتنيا ی هيلا تماخض اب ار

دي نيرت یلخاد ) یا یارب ماگنھدوز یتسيز رگناشن کي ناونع هب ديتوراک گرخرس (هيلا

ميئامن یسررب ار نيئارش بلصت یراميب .

یاھديتپپ فلتخم عاونا یناوارف یور رب یلصفم مسيتامور یراميب اب طبترم یطيحم تارطخ ريثات یسررب همان ناياپ نيا موس هعلاطم فدھ مھ و رگ شنک ی هنيلورتيس دشاب یم ديئتامور روتکاف یاھ پوتوزيا روضح نينچ

.

جياتن و اھ شور :

هرامش هعلاطم ١

دروم عون زا گرزب یتيعمج ی هعلاطم /قيقحت کي تروص هب هعلاطم نيا- -

ريگ همھ نامزاس رد یدھاش

ملھکتسا رھش یبلق یراميب یسانش

تفرگ ماجنا هرامش نيکولرتنيا لولحم هدنريگ نيئتورپ ود تظلغ ام ،قيقحت نيا رد. ۶

(SIL6R) ی هرامش نيئتورپ وکيلگ و ١٣٠

(SGP130) ر

ا زا شيب رد

رازھ هنومن رس یبلق هتکس هب لاتبم ناراميب م لرتنک و

ميدرک یريگ هزادنا ناشياھ .

تظلغ یلااب هرامش نيکولرتنيا لولحم هدنريگ نيئتورپ ۶

اب رطخ شھاک اب (SGP130)١٣٠ ی هرامش نيئتورپ وکيلگ یلااب رايسب تظلغ رگيد یوس زا و تسا طبترم یبلق هتکس رطخ شيازفا یبلق هتکس

هارمھ تسا دشاب یم یتسيز رگناشن ود نيا نيب یشيازفا مھ ی هطبار عون کي دوجو رگنايب ام جياتن. .

مش نيئتورپ وکيلگ نيياپ یاھ تظلغ ،یترابع هب ی هرا

١٣٠ (SGP130) هرامش نيکولرتنيا لولحم ی هدنريگ یلااب یاھ تظلغ و

۶ عوقو رطخ شيازفا هب رجنم یشيازفا مھ ی هطبار یعون یط ،

دنوش یم یبلق هتکس .

هرامش هعلاطم ٢

رگن هدنيآ ی هژورپ رد مود هعلاطم- IMPROVE

دنچ نامزمھ هک روشک

اپورا فلتخم یي دش ماجنا ،دنراد تکرش نآ رد

. رطخ اب نآ تارييغت و ديئتوراک گرخرس یايدما ميتنيای هيلا تماخض صخاش نيب طابترا یسررب قيقحت نيا فدھ و یبلق ضراوع زورب

دشاب یم هدنيآ رد یقورع .مونژ یلاوت قيقحت نيا رد (genomic DNA)

شور ود زا هدافتسا اب هعلاطم دروم دارفا bech

Cardiometa و

Immune chip دش نييعت

. ١٣٠ ی هرامش نيئتورپ وکيلگ تظلغ اب١٠ هرامش موزومورک زاrs1929666شخب و٣ هرامش موزومورکrs10935473 شخب ینعم یگتسباو

دنداد ناشن یراد . ٢۴ ،نينچمھ ی هرامش نيئتورپ وکيلگ تظلغ اب یديئوئلکون کت یتخير دنچ دروم ١٣٠

.دنتشاد یگتسباو هعلاطم دروم دارفا مرس رد

شخب rs17688225 هرامش یموزومورک

١۴ ی هرامش نيئتورپ وکيلگ تظلغ ثعاب زين ١٣٠

زا و تبثم یگتسباو هعلاطم دروم دارفا مرس رد

ماميتنيا ی هيلا تماخض اب یفرط د

گرخرس یاي ديتوراک

تشاد یفنم یگتسباو ی

. ١٣٠ ی هرامش نيئتورپ وکيلگتظلغ هک دھد یم ناشن ام جياتن .دوش یم ميظنت و یھدناماس یناسنا مونژ رد فلتخم یکيتنژ هاگياج دنچ طسوت

لاامتحا یکيتنژ یاھ هاگياج نيا هک تسا رکذ هب مزلا اب

ماميتنيا ی هيلا تماخض هدننک ميظنت یکيتنژ یاھ هاگياج د

يئتوراک گرخرس یاي

د .دنراد یناشوپمھ

هرامش هعلاطم ٣

دشاب یم یلصفم مسيتامور هب لاتبم یدئوس ناراميب زا هورگ ود یاھ هداد ساسا رب موس هعلاطم یاھ ليلحت- . ٢٨۵٩ دادعت لماش هکEIRA هب موسوم یلصفم مسيتامور یسانش ريگ همھ تاعلاطم هورگ یلصفم مسيتامور هب لاتبم ناراميب هورگ و هدوب راميب

قطنم سردوز å

Ume ) å Ume - (e RA ش هک دادعت لما ١٠١١ دنشاب یم رفن .

ناراميب یاھ هنومن رد یلصفم مسيتامور یراميب یصاصتخا یاھ نتداپ فلتخم عاونا زا یرايسب ی هدرتسگ و نامزمھ روضح رگنايب ام جياتن دشاب یم اھپوت یپا لباقتم شنکاو اي عيزوت و شرتسگ دنيآرف هجيتن لاامتحا اھ نتداپ نيا روضح .دشاب یم .

ا یسررب ناشن دوجوم یاھ نتداپ عاون

لامعتسا هک دھد یم

دض یاھ نتداپ روضح اب تايناخد ߙଷ଺ିହ଴

Fib - 1 , Cit - CEP عون نيبولگونوميا نينچمھ و A

دض RF تسا هتسباو

. زا یصاخ عون روضح اب(BMI) ندب ی هدوت ی هيامن شيازفا و یلکلا تابورشم مک فرصم نايمیطابترا هنوگ چيھ هعلاطم نيا رد یاھ نتداپ

LIST OF SCIENTIFIC PAPERS

I. Moreno Velasquez I*, Golabkesh Z*, Kallberg H, Leander K, de Faire U, Gigante B.

Circulating levels of interleukin 6 soluble receptor and its natural antagonist, sgp130, and the risk of myocardial infarction.

Atherosclerosis. 2015;240(2):477-81.

II. Bonomi A, Veglia F, Baldassarre D, Strawbridge RJ, Golabkesh Z, Sennblad B, Leander K, Smit AJ, Giral P, Humphries S, Tremoli E, Hamsten A, Faire U, Gigante B, on behalf of the IMPROVE study group

Analysis of the genetic variants associated with circulating levels of sgp130.

Results from the IMPROVE study.

Genes Immun. 2020;21(2):100-8.

III. Golabkesh Z, Frumento P, Hansson M, Lundberg K, Arkema E, Mathsson-Alm L, Jakobsson P, Holmdahl R, Skriner K, Serre G, Rönnelid J, Dahlqvist SR, Klareskog L, Alfredsson L, Brynedal B.

Manuscript

* Contributed equally

CONTENTS

1 BACKGROUND 1

1.1 CVD 1

1.1.1 CHD 2

1.2 Biomarkers for CVDs 5

1.2.1 IL-6 5

1.2.2 IL-6R, sIL-6R, gp130 (Biology of IL-6 inflammatory

signaling pathways) 5

1.2.3 sgp130 6

1.3 IL-6, sIL-6R, sgp130 and risk of CVDs 7

1.3.1 sgp130 genetic variants and risk of CVD 8

1.4 c-IMT and inflammatory biomarkers 9

1.5 RA 9

1.6 Biomarkers for RA 10

1.6.1 Rheumatoid factor (RF) 10

1.6.2 Anti-citrullinated protein/peptide antibodies (ACPAs) 10

1.7 Environmental risk factors for RA 13

2 AIM 14

2.1 Overall aim 14

2.2 Specific aims 14

3 MATERIALS AND METHODS 15

3.1 Materials 15

3.1.1 SHEEP study 15

3.1.2 IMPROVE study 17

3.1.3 EIRA study 19

3.1.4 eRA-Umeå study 20

3.2 Statistical analysis 21

3.2.1 Study I 21

3.2.2 Study II 22

3.2.3 Study III 24

4 RESULTS AND DISCUSSION 25

4.1 Study I, II 25

4.1.1 Association of sIL-6R and sgp130 with the risk of MI 25

4.1.2 Interaction analysis 26

4.1.3 Genetic variants associated with serum sgp130 levels 27

4.1.4 Association of sgp130 and c-IMT 28

4.1.5 Association of the SNPs associated with sgp130 with c-IMT 28

4.2 Study III 29

4.2.1 Correlation between ACPAs 29

5 METHODOLOGICAL CONSIDERATION 35

5.1 Strengths 35

5.2 Causality and reverse causality 35

5.3 Misclasification bias 36

5.3.1 Misclassification of exposure 36

5.3.2 Misclassification of outcome 37

5.4 Selection bias 38

5.5 Confounding bias 39

5.5.1 Residual and unmeasured confounding 40

6 CONCLUSION AND FUTURE PERSPECTIVE 41

6.1 Study I, II 41

6.2 Study III 41

7 ACKNOWLEDGMENTS 42

8 REFERENCES 44

LIST OF ABBREVIATIONS

ACR American college of rheumatology anti-CCP Anti-cyclic citrullinated peptide

ACPAs Anti-citrullinated protein/peptide antibodies AP Attributable proportion

Bif Bifurction

CAD Coronory artery disease CHD Coronary heart disease

Chr Chromosome

CI Confidence interval

c-IMT Carotid intima-media thickness

CC Common carotid

CRP C-reactive protein

CV Cardiovascular

CVDs Cardiovascular diseases CVE Cardiovascular events

EC Endothelial cell

EIRA Epidemiological Investigation of Rheumatoid Arthritis ELISA Enzyme-linked immunosorbent assay

eRA-Umeå Early rheumatoid artheritis Umeå cohort EULAR European league against rheumatism FDR False discovery rate

Gp130 Glycoprotein 130

GTEx Genotype-tissue expression GWAS Genome Wide Association Study HDL High-density lipoprotein

ICA Internal carotid artery

I_CC 1st centimeter of common carotid

Ig Immunoiglobolin

IL- Interleukin

IL-6R IL-6 receptor

IMPROVE carotid Intima Media Thickness and c-IMT PRogression and the

MDS Multidimentional scaling MI Myocardial infarction

MSD MesoScale Discovery

OR Odds ratio

RA Rheumatoid arthritis

RF Rheumatoid factor

S Synrgic index

sgp130 Soluble glycoprotein 130

SHEEP Stockholm Heart Epidemiology Program sIL-6R Soluble interleukin-6 receptor

TNF-α Tumor necrosis factor α WHO World Health Organization

1 BACKGROUND

Inflammation is a self-protection reaction of the body in response to the harmful stimuli like pathogens, if inflammation progress to a chronic state, this may cause diseases, such as atherosclerosis or rheumatoid arthritis (RA). The focus of this thesis was on inflammatory biomarkers and two chronic inflammatory diseases:

cardiovascular diseases (CVDs), particularly Myocardial Infarction (MI) and atherosclerosis, as well as RA.

A biomarker usually imputes to a distinct biological indicator such as measurable molecules in bodily fluids and can be used to assess both normal or pathogenic processes and evaluation of treatment responses. To be more specific, the risk or progression of a disease or the effect of a given treatment for a disease can be evaluated by relevant biomarkers, such as changes in expression or state of a protein. They can also be used to infer the genetic causes of diseases (1). For instance, the presence of autoantibodies in the serum of patients in a very early stage of the disease is a typical biomarker of autoimmune diseases. Autoantibodies as biomarkers can help the clinicians to diagnose the disease at an earlier stage and also classify the patient and provide adequate medications.

1.1 CVD

CVDs are of interest for researchers because they are the primary source of mortality in the world. The World Health Organization (WHO) estimates that CVDs will be the leading cause of death in all continents apart from Africa until 2030 and therefore remain a center of attention within the field. The 2016 Global Burden of Disease reported 31% of all deaths globally (17.9 million), as attributed to CVD (2).

Based on European CVD statistics 2016, despite all the development in diagnosis and treatment of CVDs, still more than 4 million deaths occur every year due to CVD across the continent (3). According to the WHO, CVDs are defined as a class of disorders that involve heart and blood vessels and include (4):

1) Coronary Heart Disease (CHD) 2) Cerebrovascular diseases 3) Peripheral arterial disease 4) Congenital heart disease 5) Rheumatic heart disease 6) Deep vein thrombosis

Risk factors contributing in the development of CVDs can be categorized as:

1) Modifiable risk factors: i.e. hypertension, smoking, raise blood glucose (dia- betes), lack of physical activity, unhealthy nutritional habits, hyperlipidemia, being overweight and obese. These risk factors can be controlled or treated, that in turn prevents CVDs and reduces the number of deaths each year caused by CVDs.

2) Non-modifiable risk factors: i.e. age, gender and family history of CVD. These actualities cannot be changed, however regular checkups for high-risk groups of individuals is recommended (5).

1.1.1 CHD

CHD is the most common type of CVD. Alternative titles for CHD are coronary artery disease (CAD), ischemic heart disease and atherosclerotic heart disease.

CHD consists of stable angina, unstable angina, MI and sudden cardiac death.

These are the clinical manifestation of atherosclerosis, a chronic inflammatory disease of the coronary arteries.

Despite the importance of common risk factors for the onset of cardiovascular events (CVEs), it is commonly suggested that more than 50% of patients suffer from CHD have none of the traditional cardiovascular (CV) risk factors (6). This implies that other nontraditional risk factors have a significant role in the develop- ment of CHD. Therefore, it is essential to investigate the non-conventional risk factors and new biomarkers in addition to traditional risk factors to predict CHD and consequently CVDs. This would enhance the chance of preventing the disease onset and thereby reducing CVDs mortality.

1.1.1.1 MI

MIs usually occur due to acute ischemia in myocardium where thrombus blocking the coronary arteries. Artery occlusion happens because of stenosis of the coronary artery by an atherosclerotic plaque that causes acute ischemia, lack of blood flow to the myocardium, which in turn leads to oxygen deprivation in the cardiac muscles.

The definition of MI has developed gradually with the increase in sensitivity and specificity of available biomarkers. The universal definition of MI was originally published in 2007 (19). According to the international task force, the updated uni- versal definition for MI (from 2014 (20)) concentrates on clinical criteria as well as biomarkers and consist of electrocardiogram changes and imaging. Acute MI is defined as myocardial cell death in response to extended myocardial ischemia.

1.1.1.2 Atherosclerosis

Atherosclerosis is a chronic and progressive inflammatory disease of the blood vessel wall and represents the biological lesion underlying atherosclerosis-related CVDs. The clinical manifestations of atherosclerosis are specific for different vas- cular beds and are related to the stability of focal atherosclerotic lesions, called plaque. Plaques usually contain loads of lipids, including cholesterol and fatty acids and white blood cells, particularly macrophages (7).

If atherosclerotic plaque forms in the coronary arteries, it is more likely to grow within the vascular wall and reducing the blood flow in cardiac tissue and conse- quently, facilitate the onset of chronic ischemia. Atherosclerotic lesions are cat- egorized as stable or unstable (vulnerable) plaques. Stability of an atherosclerotic plaque is more important than its size and depends on its cellular and extracellular contents.

Stable plaques are characterized by a thick fibrous cap, small lipid cores, few inflammatory cells and many smooth muscle cells. Stable plaques usually result in clinical symptoms described as stable angina. At the other side of the spectrum are atherosclerotic plaques in the coronary arteries, which are characterized by large lipid cores, high inflammatory content (numerous macrophages) and pro- angiogenic activities. These are called unstable plaques. They have relatively few smooth muscle cells and a very thin fibrous cap that tends to rupture in response to different stimuli (8, 9).

When the cap rupture, is followed by activation of the coagulation cascade and thrombus formation. Existing platelets and coagulation factors in blood circulation will be faced by the plaque’s extracellular matrix and lipid core that has thrombo- genic characteristics. Afterwards, the disrupted plaque is used as scaffolding for platelet aggregation and coagulation, eventually thrombus formation.

Thrombus size depends on the plaque rupture area. When it is large enough, the thrombus can occlude the coronary vessel lumen. The blockage can be either partial or complete, thereby accelerate the occurrence of an acute coronary event. The clinical manifestation of plaque rupture or superficial erosion is often MI (10, 11).

Atherosclerosis might remain subclinical during years, which indicates the require- ment of different methods to identify the presence of atherosclerosis. Available techniques such as intravascular ultrasonography and B-mode ultrasonography are used to measure the surrogate markers of atherosclerosis. According to the estimations, the different dosage of statin treatment can be prescribed to reduce, cease or even inverse the progression of atherosclerotic disease (12).

Atherosclerosis and inflammation

Several lines of evidence indicated the key role of inflammation in all phase of atherosclerosis. Various cytokines like IL-12, IL-6, IL-1, IL-2, IL-8, IL-10, IL-17, tumor necrosis factor α (TNF-α), and C-reactive protein (CRP) are associated with CVDs and involved in the inflammatory processes contributing to atherosclerosis and plaque formation in the vessel walls of human (13). However, the independent contribution of inflammatory biomarkers to the risk of CVD is still not fully known.

Subclinical atherosclerosis

Carotid intima-media thickness (c-IMT) is a surrogate marker for early athero- sclerosis that reflects the extension of the thickness of the intima-media layer in carotid artery. c-IMT is used as an intermediate phenotype for diagnosis of early stages of atherosclerosis. Several large population-based studies have indicated c-IMT as a valid estimation of the risk for CVEs (14).

Based on the results of different epidemiological studies, c-IMT is associated with various CV risk factors like diabetes and impaired glucose tolerance, hyper- cholesterolemia (15, 16), high-density lipoprotein cholesterol and triglycerides (17).

Using c-IMT as a biomarker of CVD is beneficial because it is a non-invasive, un-complicated and a low-cost method. Moreover, it is reproducible and radiation- free. This test can classify individuals as high or low risk groups for CVDs, and appropriate precautionary action plans can be implemented accordingly (12).

c-IMT usually is measured in far and near walls of four segments of both right and left carotid arteries: at the internal carotid arteries, at the carotid bifurcation, in the 1st cm of the common carotid, in the proximal artery (18).

Differences in c-IMT in different parts of the artery can reflect the variation in the local hemodynamic forces. Usually, c-IMT measurement of the deeper wall (far wall) is more reliable for diagnosis of atherosclerosis compared to the adjacent wall and is thus the most applied in clinical studies.

Based on guidelines presented jointly by European society of hypertension and European society of cardiology, the normal values for c-IMT is <0.9 mm, and c-IMT value between 0.9 mm and 1.5 mm is an indication of atherosclerosis and high risk of CVD. c-IMT values >1.5 mm demonstrate asymptomatic carotid plaque.

Changes in c-IMT can reflect regression and or progression of atherosclerotic CVD (18). If the increased in c-IMT is detected in the early stage, there is a pos- sibility of prevention or treatment before more serious damage to the vessel wall.

1.2 BIOMARKERS FOR CVDS 1.2.1 IL-6

IL-6 is a 26-kDa inflammatory cytokine with a four-helix bundle structure and consists of 184 amino acids (21). IL-6 is a pleiotropic cytokine that has pro-inflam- matory as well as anti-inflammatory characters. This key cytokine has different effects on the immune system, liver, the endothelium and the vascular smooth cells. IL-6 influences the accumulation of platelets and proliferation of vascular smooth muscle cells in coronary arteries (22). In hepatocytes, IL-6 regulates the production and the release of the acute-phase proteins, for instance, CRP (23).

In endothelial cells, IL-6 has a regulatory effect on the expression of adhesion molecules as well as other cytokines such as IL-1β and TNF-α. The secretion of these cytokines, along with IL-6 from endothelial cells creates a microenvironment that facilitates the initiation of an inflammatory process in the vessel wall. This represents one of the central mechanisms involved in the onset and progression of atherosclerosis (24).

1.2.2 IL-6R, sIL-6R, gp130 (Biology of IL-6 inflammatory signaling pathways)

IL-6 mediates its biological activities by binding to two membrane proteins: an 80-kDa glycoprotein called IL-6 Receptor (IL-6R) and a 130-kDa trans-membrane signal transducer protein named glycoprotein130 (gp130). IL-6R is expressed on leukocytes, monocytes and hepatocytes, where it regulates the release of acute-phase proteins and inflammatory cytokines (25). The soluble form of IL-6R (sIL-6R) is present in serum (ng/ml) and is a 50-55 kDa protein lacking the trans-membrane and cytoplasmic regions of IL-6R. sIL-6R is produced either by limited proteolytic shedding of the membrane-bound receptor or by translation from alternatively spliced mRNA (26) (Figure 2).

The IL-6 signaling pathway that is activated in the presence of membrane bound IL-6 receptor is called classic-signaling, which mostly regulate the beneficial regenerative and anti-bacterial roles of IL-6. The alternative signaling pathway that is mediated by soluble IL-6R is known as trans-signaling. It is suggested that trans-signaling is accounted for the majority of the detrimental IL-6 effects (27).

In both signaling pathways, IL-6/IL-6R or IL-6/sIL-6R complex is followed by the homodimerization of gp130 subunits that are naturally expressed on all cell types. Thereafter the intercellular signaling cascade is initiated (Figure 1).

gp130 has no measurable affinity for IL-6 or IL-6R alone, but a high affinity for both IL-6/IL-6R and IL-6/sIL-6R complexes. Therefore, all cells that have gp130 present in their membranes can be responsive to IL-6, even in the absence of IL-6R, by the complex of IL-6 and sIL-6R (28).

1.2.3 sgp130

A soluble form of gp130 (sgp130) exists in human blood in three isoforms (50- 110kDa); sgp130-RAPS, sgp130-E10 and full-length sgp130 (29-31). They are produced by proteolytic cleavage of gp130 and or alternative splicing of mRNA encoding the transmembrane protein (32). sgp130 represents a natural antagonist of the IL-6/sIL-6R pathway (33). sgp130 inhibits the trans-signaling through specific binding to the IL-6/sIL-6R complex with a high affinity (1mM). IL-6/

sIL-6R complex coordinates the pro-inflammatory and the proatherogenic effects of IL-6 through trans-signaling (34, 35) and based on several animal and in vitro studies sgp130 circulation can prevent its binding to the membrane-bound gp130 Figure 1. IL-6 classic-signaling pathway through IL-6R (left panel) and trans-signaling pathway through sIL-6R (right panel).

+ IL-6 classic

signaling pathway

IL-6R

JAK JAK

MAP PI3K

AKT

IL-6

gp130 gp130 IL-6R

Cytoplasm Extracellular

IL-6 gene expression

sIL-6R sIL-6R

IL-6 trans- signaling pathway

IL-6

gp130 gp130

STAT3 P

STAT1 P

JAK JAK

IL-6

IL-6 IL-6

IL-6 gp130 gp130

thus can be an ideal pharmacological target for IL-6 signaling cause inflamma- tory conditions. According to an in-vitro study, sgp130 play an inhibitory role for IL-6 trans-signaling mostly in the full-length isoform (38). Note that it is not possible to recognize which isoform of sgp130 is detected by biological assays.

The potential therapeutic application of sgp130 to prevent CVDs is of interest for several research institutions.

sIL-6R

C: IL-6 trans-signaling

gp130  gp130 

STAT3 STAT1 P P

JAK JAK

IL-6

PM

Nucleus sgp130 D: IL-6 trans-signaling

gp130 gp130IL-6

Antagonist Agonist

Pro-Inflammtory

Increase EC adhesiveness CRP↑

Angiogenesis Inflammation

sIL-6R

mIL-6R

ADAM17 A: Proteolytic shedding

sIL-6R

B: Alternative mRNA splicing IL-6R gene

mIL-6R

Anti-Inflammtory

sIL-6R

Figure 2. The IL-6 signaling pathways: soluble forms of IL-6R and gp130 are produced by proteolytic cleavage (A) and or alternative splicing (B) of the full-length membrane- bound forms. The pro-inflammatory effect of IL-6 occurs through trans-signaling (C) and sgp130 can inhibit the trans-signaling (D).

1.3 IL-6, SIL-6R, SGP130 AND RISK OF CVDS

IL-6 is a known and independent risk indicator of various CVDs. In particular, increased serum level of IL-6 is associated with elevated MI risk (39). One pro- spective study of MI showed men with high IL-6 plasma levels had 2.3 times higher risk of developing MI compare to those with low levels, still significant after adjusting for traditional CV risk factors (40). Moreover, results from two

sIL-6R seems to mediate the inflammatory effects of IL-6 via trans-signaling (42) (Figure 3). The role of IL-6 trans-signaling pathway (through IL-6/sIL-6R complex) has been emphasized in retaining chronic inflammation in systemic dis- eases like atherosclerosis (43). In a case-control study, patients with MI and CHD showed an elevated level of sIL-6R compared to controls, in the same population no difference in level of sgp130 was observed in cases and controls (44, 45). In other studies of the heart failure, an increased level of sgp130 has been reported in patients (46-48). The causal effect of IL-6 trans-signaling on development of atherosclerosis, as a key cause of CVDs, has been investigated by different clini- cal (49) and experimental studies (50, 51).

It has been studied if blocking IL-6 trans-signaling can prevent the inflammatory effects of IL-6 in different diseases. In this regard, a recombinant form of sgp130 (sgp130Fc) has been applied as IL-6 trans-signaling inhibitor in several inflam- matory and degenerative disease models in the animal and in-vitro studies (52, 53). An experimental research performed on a mouse model indicated that the administration of sgp130Fc has a significant protective effect on athero sclerosis.

It showed that sgp130 reduced the atherosclerotic plaque development and pro- gression through nullifying the IL-6/sIL-6R trans-signaling (50). This suggests that sgp130 may antagonize the effect of IL-6/sIL-6R complex on the progression of atherosclerotic lesions. sgp130Fc is in fact now tested as an anti-inflammatory drug in phase 1 clinical trials for rheumatoid arthritis (54).

1.3.1 sgp130 genetic variants and risk of CVD

Genetic and environmental factors together define if an individual is susceptible to CVD. Genome Wide Association Study (GWAS) have provided a platform to explore the association between specific genetic variants and diseases.

The genes which regulate the level of sgp130 mainly have not been discovered yet. The association of polymorphisms in the gp130 gene with CVD risk has been investigated in a few studies. One single nucleotid polymorphism (SNP) (rs3729960) in GP130 at chromosome five is associated with a decline in the risk of MI, independently of other conventional CV risk factors (55). Furthermore, one study on two separate population, 546 subjects from OSLO high-risk population for CAD and 299 subjects from VIENA population with angiographically proven CAD, demonstrated a higher level of sgp130 in the individuals carrying the gp130 polymorphism G148C in comparison with wild type (56).

Increasing the knowledge on genetic variants regulating sgp130 is important, to understand the mechanisms underlying generation of sgp130 as an inhibitor of IL-6

1.4 C-IMT AND INFLAMMATORY BIOMARKERS

Research on the association between inflammatory biomarkers and c-IMT as an indicator of CVD is essential for more precise risk prediction of CVDs as well as for effective prevention and treatment of atherosclerosis. Some previous studies have shown the association between c-IMT and inflammatory biomarkers. Wang et al.

have indicated the association between c-IMT and CRP in women (57). Furthermore, an independent positive association between IL-6 level and c-IMT has been reported in patients with growth hormone deficiency who are at higher risk of CVDs (58).

Moreover, results from one study in patients with obstructive sleep apnea has shown significant correlation between c-IMT and CRP, IL-6 and IL-18 serum levels. This proposes an association between systemic inflammatory biomarkers and elevated risk of atherosclerosis progression, thus increasing the risk of CVDs in patients with obstructive sleep apnea (59). Another finding from an epidemiological study with individuals at risk of CVDs (having at least three different traditional CV risk factors) showed an inverse association between the plasma IL-5 concentration and changes in c-IMT over 30 months follow-up in women (60).

1.5 RA

RA is the most common chronic inflammatory autoimmune disease, and a conse- quence of the faulty attacks of the immune system to the body tissue, specifically synovial tissue. The result is inflammation and stiffness of the joint linings, and unless successfully treated, it leads to bone erosion and loss. RA initially affects the wrist and small joints of the hands, such as middle joints of the fingers, and typically leads to painful and swollen joints (61). The disease involves symmetric pattern in pain, stiffness and swelling in peripheral joints on both sides of the body. Other symptoms such as fever, fatigue or low energy and reduced appetite can appear gradually. It can take weeks and months to verified diagnosis (62-64).

The inflammation might expand to other parts of the body like lungs, heart, eyes and nerves. The cause of RA, like other autoimmune diseases is complex and involving both genetic and environmental factors (65). RA can appear in any age, but its incidence rate increases with age (up to age 55-80) and women have 2.5-3 times higher risk. It is estimated that RA affects nearly 0.5-1% of the whole popu- lation of the world (66). According to data from nationwide register-based study and Swedish Rheumatology Quality register in Sweden, RA incidence is 41 per 100,000 (67, 68). Various inflammatory cytokines like IL-6, TNF-α and IL-1 are highly expressed in RA. TNF-α is the main cytokine causing inflammation in the synovium of RA patients. Increased level of IL-6 is measurable in synovial fluids of RA patients with high disease activity. Synovial membrane express IL-6, and there is a correlation between the level of IL-6 and the level of radiological joint

1.6 BIOMARKERS FOR RA 1.6.1 Rheumatoid factor (RF)

RF is an autoantibody that was originally found in RA patients. IgM-RFs are the most observed isotype, but IgA, IgG, IgE, IgD RFs may also be detected. RFs have been reported in other autoimmune and non-autoimmune diseases as well as in healthy individuals (69). The RF test has high sensitivity but low specificity for RA. RF is detected in about 75% of the RA patients, but it is also present in other autoimmune diseases and infectious diseases. Positive RF test is not sufficient for RA diagnosis but aids diagnosis. For being classified as a RA patient according to the 1987 American College of Rheumatology (ACR), the patients must fulfill at least four of seven criteria (Table 1), where RF positivity is one criterion.

Table 1. The 1987 revised criteria for the classification of RA*

The 1987 revised criteria for the classification of RA*

1.Morning stiffness

2. Arthritis of 3 or more joint areas 3. Arthritis of hand joints

4. Symmetric arthritis 5. Rheumatoid nodules 6. Serum rheumatoid factor 7. Radiographic changes

* The patients must have at least 4 of these 7 criteria to be classified as RA case

Source: The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis (70)

1.6.2 Anti-citrullinated protein/peptide antibodies (ACPAs)

Citrullination is a posttranslational protein modification where the unusual amino acid citrulline substitutes an arginine. Calcium-dependent peptidyl arginine deimi- nase (PAD) is an evolutionarily conserved protein that catalyzes the citrullination of proteins. Several isoforms for PAD are known in human (PAD1–4 and PAD6).

PAD can citrullinate the extracellular proteins such as collagen and fibrinogen in tissues under inflammatory conditions (71).

Antibodies directed against citrullinated peptides are RA specific (72-74). They are routinely measurable using an anti-cyclic citrullinated peptide (anti-CCP) test. The recent generation of CCP (CCP-2) test is an ELISA based assay that captures reactivity towards either in vitro or synthetic citrullinated peptides. It is

ACPAs are predictive marker used by clinicians for early RA diagnosis. The updated diagnostic criteria from 2010 from ACR and EULAR are shown in Table 2 (72). It is more likely that RF positive patients are also ACPA positive. Since anti-citrulline immunity and RF autoantibodies are expressed even years before the onset of any symptoms, it is believed that they are involved in the early process of joint destruction and the pathogenesis of RA.

Data from a longitudinal study showed that 75% of 318 undifferentiated inflam- matory arthritis subjects with anti-CCP positive at baseline have classified as RA patients after one-year follow-up. This number increased to 93% after three years of follow-up. Meanwhile, only 25% of anti-CCP negative individuals had progressed to RA after three years. A similar study of 314 early RA patients reported that 90% of anti-CCP positive participants were classified as RA after one-year follow-up (75).

This emphasizes on prognostic characteristics of anti-CCP antibodies as a biomarker for RA and can help patients to be prioritized for treatment.

Importantly, different RA patients display reactivity towards different specific citrul- linated peptides (65, 76). Commercial CCP-2 tests are meant to identify any reac tivity towards citrullinated peptides, but they do not specify which certain citrullinated peptides that the individual has reactivity against.

Table 2. The 2010 ACR/EULAR criteria for the classification of RA. ESR: Erythrocyte sedimentation rate.

Classification criteria for RA (score-based algorithm: add scores of categories A-D:

a score of ≥6/10 is needed for classification of a patient as having definite RA Score A. Joint involvement

1 large joint 2-10 large joints

1-3 small joints (with or without involvement of large joints) 4-10 small joints (with or without involvement of large joints)

>10 joints (at least 1 small joint)

0 1 2 3 5 B. Serology (at least 1 result is needed for classification)

Negative RF and negative ACPA Low-positive RF or low-positive ACPA High-positive RF or high-positive ACPA

0 2 3 C. Acute phase reactants (at least 1 test result is needed for classification)

Normal CRP and normal ESR Abnormal CRP or abnormal ESR

0 1 D. Duration of symptoms

<6 weeks 0

Recently a new multiplexed array has been developed that simultaneously captures reactivity towards multiple citrullinated antigens. This array includes more than 40 citrullinated autoantigens that were subdivided according to their “protein of origin”. Table 3 summarizes the citrullinated peptide antigens on the multi plex chip array. Citrullinated peptide antigens on the multiplex microarray were named based on their protein source and the location of amino acid in the peptide chain that arginine residue was substituted by citrulline residue.

Table 3. Citrullinated peptide antigens on the multiplex microarray.

Protein source Citrullinated Peptide

Fibrinogen cit-Fibß36-52, cit-Fibα563-583, cit-Fibα580-600, cit-Fibα621-635, cit-Fibα36-50, cit-Fibß60-74, cit-Fibßcit-R, cit-FibßR-cit, CCP-1

Vimentin cit-Vim60-75, cit-Vim2-17

α-enolase CEP-1

Collagen type II cit-C1, cit-F4R-cit, cit-F4cit-cit, cit-F4cit-R

filaggrin CCP-1

hnRNP A3 cit-Pept-Z1, cit-Pept-Z2, cit-Pept-1, cit-Pept-5, cit-Pept-Bla26

Peptides on the array included in study III derived from:

1) vimentin, which is fibroblast intermediate filament found in the cytoskeleton in eukaryotic cells and present in around two-third of RA patients (78).

2) fibrinogen is a hexameric plasma glycoprotein central to coagulation. Around 50-60% of RA patients have immune reactivity towards citrullinated fibrinogen epitopes (79).

3) alpha-Enolase (α-Enolase): is a ubiquitously expressed glycolytic enzyme.

Around 40% of RA patients express antibody against citrullinated-peptide epitope CEP-1, which is an immunodominant epitope in this group of peptides (80).

4) collagen type II are the triple-helical peptides that are the main component of hyaline cartilage. Anti-citrullinated collagen type II antibodies are found in about 40% of patients with RA (81).

5) filaggrin is a filament-associated protein that attaches to keratin fibers, mostly found in epithelial cells.

6) Heterogeneous Ribonuclear Protein A3 (hnRNP A3).

The presence of processes like cross-reactivity and epitope spreading among ACPA are widely known, which contribute to the co-occurrence of these RA specific citrullinated peptide antibodies. (82, 83).

The ability of an antibody to react with a similar epitope on different antigens call cross-reactivity. Some of the antibodies towards citrullinated peptides may react to more specific citrullinated antigens, but most of them tend to be widely cross-reactive (82, 83).

Epitope spreading is characterized as the development of immunity against diverse epitopes. This immune response can be directed towards a self or foreign antigen, to subdominant epitopes on that protein instead of the immunity reactivity against dominant specific epitope (84, 85). Co-occurrence of ACPAs from epidemiological aspect can be translated to mediating or colliding effect, meaning each antibody can mediate the presence of another antibody.

1.7 ENVIRONMENTAL RISK FACTORS FOR RA

Environmental factors can play an essential role in the origination of systemic inflammation and autoimmunity of RA prior to visible joint symptoms.

Smoking is a well-established exposure that associates with elevated risk of ACPA positive RA but not ACPA negative RA. It has been shown that smoking cause citrullination in the lung and has an association with the occurrence of anti- citrulline immunity in RA (86-88). Alcohol consumption has been shown to have a significant inverse association with risk of RA (89-93). A higher body mass index is associated with increased risk of ACPA negative RA (93-95). It is not known yet if RA risk factors have an association with specific anti-citrullinated anti bodies and thereby can be potentially engaged in RA pathogenesis through inducing anti-citrulline immunity.

As it was mentioned, the effects of environmental risk factors differ between ACPA positive and ACPA negative RA patients (66). ACPA positive RA patients also show a more severe course of the disease compared to ACAP negative RA patients (96).

Consequently, distinct phenotypic manifestations of RA have an association with distinct RA relevant risk profile. This implies ACPA positive and ACPA negative RA patients go through different pathogenic mechanisms to develop the disease.

It is suggested that these antibodies, and the specific immunity they represent, are involved in the pathogenesis of RA. Since they are expressed many years before the onset of the disease, they have the prognostic ability.

Increasing the knowledge in specific anti-citrulline immunity in association with

2 AIM

2.1 OVERALL AIM

The aim of this thesis was to evaluate the role of new biomarkers in chronic inflammatory diseases focusing partly on atherosclerosis and its primary clinical manifestation MI, and RA.

2.2 SPECIFIC AIMS

The specific aims of this thesis were:

• To investigate the association between sIL-6R and sgp130, respectively, with MI and explore the potential interaction between these biomarkers.

• To explore the association between sgp130 serum levels and c-IMT and c-IMT progression after 30 months follow-up in a European population with high-risk profile for CVDs.

• To identify genetic variants associated with sgp130 serum levels and to inves- tigate if those genetic loci are associated with c-IMT as a marker of subclinical atherosclerosis.

• To identify if environmental risk factors of RA are associated with the presence of RA specific autoantibodies.

3 MATERIALS AND METHODS

3.1 MATERIALS 3.1.1 SHEEP study

Study I was based on data from the Stockholm Heart Epidemiology Program (SHEEP), which is a large population-based case-control research designed to investigate genetic and environmental risk factors of relevance for the occurrence of MI. The study base was consisted of all Swedish citizens living in Stockholm county from 1992-1994, aged 45 to 70 without previous clinically diagnosed MI.

3.1.1.1 Case identification and control selection

Overall, 5452 participants were enrolled in the study, 2246 cases and 3206 controls.

The cases were included in the SHEEP at the time of disease incidence at ten emer- gency hospitals within the Stockholm county. Cases were diagnosed according to the diagnosis criteria for MI, accepted by the Swedish Association of Cardiologists in 1991-1994. In this thesis, only non-fatal cases (n=1213) were included that are patients who survived at least 28 days post-MI and had no further MI before sample collection. Fatal cases (n=603) were excluded due to the absence of blood sampling.

Controls (n=1561) who matched on sex, age (five years interval) and hospital catch- ment area were randomly selected from the study base within two days from case incidence. In order to substitute potentially nonresponsive controls, five controls per case were selected simultaneously. Occasionally both the first and potential substitute controls were contacted and included in the study due to the late response of the initial control. Therefore, there are more controls than cases in the study.

Both cases and controls candidates were identified through the National Patient Registry and were investigated if they had previous MI (ICD9-codes 410 or ICD- 10 I21).

Clinical investigations were undertaken on average three months after the index events, including blood samplings under fasting conditions with the collection of whole blood for DNA extraction, serum and plasma (EDTA and citrated). The serum samples were kept at -70°C until analyzed. To collect data on a large set of potential risk factors, cases and controls completed a questionnaire, which was complemented with a telephone interview, to complete missing information. Controls underwent a health examination as close as possible in time to the corresponding cases to avoid biases due to seasonal change in blood parameters. Non-fatal cases and controls had participation rate of 87% (n=1643) and 73% (n=2339) respectively.

Anthropometric measurements such as BMI (Kg/m²), systolic and diastolic blood pressures were evaluated. Other biochemicals were also measured such as insulin, total cholesterol, triglycerides, HDL cholesterol, CRP, serum glucose, blood lipids and TNF-α.

3.1.1.2 Biomarker measurement sIL-6R

The serum level of sIL-6R was assessed in 1785 serum samples from 682 cases and 1103 controls using MesoScale Discovery (MSD) Human Cytokine assay (Gaithersburg, MD, USA), following the manufacturer’s assay protocol. Samples were diluted 1:75, and the calculated concentrations from the standard curve were expressed in (ng/mL). The minimum detectable value for sIL-6R was 0.1 pg/mL.

The intra-assay variability was asessed by running n=267 samples in duplicate within the same experiments, whereas the inter-assay variability was assessed by duplicating n=67 samples in separate experiments. The intra-and inter-assay coef- ficient variations were 6.1 % and 3.8% respectively. According to the manufacturer, the recommended threshold is 15% and 18% for mean intra-assay and mean inter- assay coefficient of variation, respectively.

sgp130

The serum level of sgp130 were measured using an assay from R&D systems®, Quantikine® ELISA according to the protocol instructions. Because of lack and or inadequacy of serum, in total, 1726 serum samples were evaluated for sgp130, 664 cases and 1062 controls. Samples were diluted 100 times, and serum concentrations (ng/mL) were derived from the standard curve. The intra variability (1.8%) and inter variability (12.1%) were calculated respectively by duplicating n=25 samples within a plate and n=37 samples in independent experiments. No specific threshold for intra- and inter-assay variability were suggested by the manufacturer for sgp130, although previous studies have reported an intra-assay variability <10–11% and inter-assay variability <10–16% (97, 98).

3.1.1.3 Ethical consideration

The SHEEP study was conducted in accordance with the Helsinki Declaration and was approved in 1991 by the Regional Ethical Review Board at Karolinska Institutet. Participants gave their oral informed consent since no written informed consent was in use.

3.1.2 IMPROVE study

IMPROVE (carotid Intima-Media Thickness and c-IMT PRogression and the risk of Vascular Events) is a multicenter prospective study designed to investigate the association of c-IMT and c-IMT progression with the risk of future CVEs. From March 2004 to April 2005, 3711 subjects (men, N= 1772 and women, N=1931) aged from 55 to 79 years free of any CVDs but with medical history for at least three conventional CV risk factors (e.g. hypertension, diabetes, dyslipidemia, smoking and family history of CVDs) were selected by seven recruiting centers from five European countries: France (n=501 from Paris), Sweden (n=533 from Stockholm), Finland (n=1050 from two centers in Kuopio), Italy (n=1095 from Milan and Perugia) and the Netherlands (n=532 from Groningen).

At baseline, every study participant completed an extensive questionnaire on lifestyle habits, former disease and treatment. Anthropometric measures were recorded, and a large biobank with whole blood, serum and plasma was established and stored at –80°C until used. For evaluating c-IMT changes overtime, subjects were followed for three years, and ultrasonographic measurements were repeated at two time-points, at 15 months and 30 months, using the same ultrasonographic protocol applied at baseline. Smoking was defined as current smoking. Hypertension was defined if self- reported and or diastolic blood pressure ≥90 mmHg and or systolic blood pressure

≥140 mmHg and or treatment with antihypertensive drugs; Diabetes was defined as self-reported and or blood glucose level ≥7 mmol/L and or treat ment with insulin or oral hypoglycemic drugs. Hypercholesterolemia was defined as LDL cholesterol

≥4.13 mmol/L and or treatment with cholesterol-lowering drugs. More details about study design have been described elsewhere (99).

3.1.2.1 Ultrasonographic measures

Participants underwent c-IMT measurements by trained sonographers using a Technos system (Esaote, Genova, Italy), equipped with a 5-10 Mhz linear array probe. Ultrasounds data were collected from the far walls of the left and right common carotid (CC), the bifurcation (Bif), the internal carotid artery (ICA) and the 1st centimeter of common carotid (I_CC) in anterior, lateral, and pos- terior angles. All measurements were done in at least three different frames at three time-points: at baseline, after 15 months and after 30 months of follow-up.

The baseline c-IMT ultrasonographic measurements selected for the study II were:

1) IMTmean: the average of mean c-IMT for all the eight segments (left and right I_CC, CC, Bif and ICA); 2) IMTmax: the largest c-IMT value recorded among all eight the segments investigated; 3) IMTmean-max: the average of the eight max c-IMT values recorded at each of the eight segments. c-IMT baseline values are reported in mm.

c-IMT was also measured after 15 and 30 months of follow-up and the progression of c-IMT was calculated at 15 months, by dividing the difference between the 15 months and corresponding baseline value by the length of intervening time period.

c-IMT progression at 30 months calculated by linear regression model between three-point measurements and expressed in mm/year.

3.1.2.2 Measurement of serum sgp130 levels

Serum sgp130 levels were measured by DuoSet ELISA development kits of human sgp130 (DY228) provided by R&D Systems ® (R&D systems Minneapolis, MN, USA). Samples required a 100-fold dilution and the range of a standard curve was 20 ng/mL to 0.25 ng/mL. Briefly, 96-well plates were coated by diluted capture antibody in the working concentration of 4.0 μg/mL sealed and incubated overnight at room temperature. Then for blocking step, 2% Bovamin serum albumin was used (200 μL/well) and incubated for one hour at room temperature. Detection antibody in working concentration of 0.08 μg/mL was added and incubated for 2 hours at room temperature. To optimize the protocol for serum samples, differ- ent concentrations for capture antibody and detection antibody were tested. To validate the sample diluent (0.2% BSA in 1X PBS), linearity test was performed by adding (spiking) known amount of human recombinant sgp130 to the samples.

Microplate reader set to 450 nm and a correction wavelength of 540 nm or 570 nm.

To calculate the intra- and inter-assay coefficient variation, a known concentration of 5 ng/uL from recombinant sgp130 was duplicated in both the same and two different plates. The intra- and inter-assay coefficients of variation were 1.88%

and 12.1% respectively.

3.1.2.3 Genotyping; CardioMetaboChip 200k and the ImmunoChip Genomic DNA from IMPROVE study participants was genotyped using two custom-made genotyping arrays; 1) CardioMetaboChip 200k: a custom Illumina iSelect genotyping array for the study of genetic variants associated with meta- bolic and CVDs, and 2) ImmunoChip: is a custom Illumina Infinium HD array containing approximately 200,000 variants mapping in genetic regions identified by GWAS as potentially relevant for immune-mediated diseases. More detailed information on these two arrays can be found (100, 101).

3.1.2.4 Ethical consideration

The IMPROVE study was funded by the Vth European Union program. The study was carried out in accordance with the Helsinki Declaration and approved by the IRB at each one of the seven recruiting centers: 1) the Regional Ethics

Ca´Granda, Milano, both in Italy, 5) the IRB at the University Hospital Groningen, Groningen, the Netherlands, 6) the IRB Hospital District of Northern Savo and 7) and the IRB at University of Eastern Finland, both in Kuopio, Finland. Each par- ticipant provided two different written consents one for general participation in the study and one for genotyping.

3.1.3 EIRA study

The EIRA (Epidemiological Investigation of Rheumatoid Arthritis) is an ongoing population-based case-control study since 1996 recruited subjects 18-79 years aged from defined (southern/central) regions of Sweden. Only cases, who were selected from May 1996 until November 2009, were included for the analysis in this study.

The participation rate for the cases was 94%.

3.1.3.1 Case identification and control selection

Cases (n=2859) were defined as RA patients diagnosed according to the 1987 ACR criteria by rheumatologists within twelve months after the onset of joint disease symp- toms. Sampling was done at the first visit before applying any RA specific medication.

Controls (n=581) were randomly selected through the Swedish national register and matched for age, sex and residential area, more details on study population can be found in (102). Data derived from the questionnaire on lifestyle-related risk factors and blood samples of all participants were collected at baseline for further genetic and serological analysis.

Subjects were defined regarding exposure to smoking to two categories: “smokers”

and “never-smokers”. Smokers are individuals that presently smoke, had smoked before, or those who occasionally smoke whereas never-smokers had not ever smoked cigarettes. Similarly, regarding alcohol intake, participants were classified either as

”ever-drinker” or ”never-drinker”. Moreover, the label ”high alcohol consumer”

was assigned to men consuming at least 168 grams of alcohol per week, and to women having at least 108 grams per week. All others were classified as low alco- hol consumers. BMI was classified as obese (≥30 kg/m²) or not obese (<30 kg/m²).

For additional analysis, two other categories were defined: 1) overweight or obese (≥25 kg/m²) and 2) normal or underweight (<25 kg/m²).

3.1.3.2 Measurement of serum level of IgG specific ACPA

A microarray-based on the ImmunoCAP ISAC system (Phadia AB, Uppsala, Sweden) was customized to measure the level of antibodies against various citrullinated pep- tides. This array has been validated using ELISA-based technology. Details were described in (76). More than 40 different citrullinated peptides and their arginine-